1. Technical Field
The present invention is related to a device of processing a dead pixel, more specifically to a device of processing a dead pixel for an image sensor.
2. Description of the Related Art
In an image sensor, a plurality of pixels are arranged in a two-dimensional structure. Each pixel converts inputted light to an electrical signal according to the luminance of the inputted light. The amount of the light inputted into each pixel can be recognized by measuring the converted electrical signal. By use of this, an image per unit pixel can be structured.
The image sensor consists of a pixel array having several hundred thousand to million pixels, a device converting analog data to digital data and several hundred to thousand storing device. There can be possibility of having an error during each pertinent process due to a large number of devices. The error causes generating a dead pixel or a hot pixel of pixels. The dead pixel becomes an important factor in the grade and pricing of the image sensor. The pixel data generated by the dead pixel is too larger or too smaller than the pixel data generated by adjacent pixels.
A conventional dead pixel processing method mainly compares a pixel with a reference pixel. If the difference between the pixel and the reference pixel is not within a predetermined range, the pixel is defined as the dead pixel and is corrected. However, the conventional dead pixel processing method uses a lot of hardware resources. In the case of a dense image or an image having high levels of the luminance difference, the difference between adjacent pixels has high levels. Accordingly, the conventional dead pixel processing method considers a proper pixel as the dead pixel in error and corrects the proper pixel. As a result, the recovered image has a lot of distortions as compared with a corresponding actual image.